Chelsea Eats serves approximately 2,000 families, most of whom are recruited from pantries across town, and gives them up to $ 400 per month on cards, depending on household size. The cards can be used anywhere that Visa payments are accepted. The program is supported by city funds, government grants and philanthropic contributions.

“This is a more efficient way to address food insecurity,” said Ambrosino. He said the program has reduced the overhead of operations like food pantries that require trucks, temporary workers and other “non-food” costs, just to get people to line up for things they may not want or need.

On Monday, researchers at Harvard Kennedy School’s Rappaport Institute for Greater Boston will publish an analysis of 47,624 transactions valued at nearly $ 2.1 million. They found that about 73 percent of those purchases were made in places where food is the primary product, such as grocery stores, wholesale clubs, convenience stores, and restaurants.

Market Basket supermarkets received 32 percent of total spend, according to the report, with most of the business going to the chain’s Chelsea store.

The program is a departure from traditional relief efforts, where participants can either be provided with food directly or only purchase items that are considered eligible.

Chelsea Eats is one of the most ambitious US experiments with so-called guaranteed income programs. The city was hardest hit by both the coronavirus pandemic and the economic crisis that followed, which left many residents out of work.

According to US census data, Chelsea has about 40,000 people in 13,000 households, which means the program has more community penetration than other such testing programs.

In Stockton, California, of 313,000 residents, a similar program that enrolled 125 people ran for two years from early 2019. Research found that participants were healthier, found better economic opportunities, and were more isolated from financial instability than their peers. They don’t get the $ 500 monthly payments.

The initial results from Chelsea suggest that the program is working as expected, according to Jeffrey Liebman, director of the Taubman Center for State and Local Government at Harvard.

Chelsea Eats participant Gloria Caballero said the program helped her afford products that other aid programs don’t cover.Matthew J Lee / Globe staff

Liebman, lead author of the paper analyzing buying patterns, said the preliminary results showed a surprising amount of money remained in and around Chelsea – in shops, restaurants, and markets – suggesting the program’s economic impact will be greater could be as expected.

While some of the restaurant spending went to big chains like McDonalds and Dunkin ‘, significant amounts went to local restaurants like Broadway House of Pizza and Taqueria El Charrito.

“What this program shows so far is that it is possible to implement a basic income program on a large scale and that it helps people meet their basic needs,” said Liebman.

Jill Shah, president of the Shah Family Foundation, which is a major supporter of the initiative, said more programs like the one in Chelsea are needed.

“People do not earn enough to provide them with enough food, shelter and security,” said Shah. “If we can’t guarantee every American that these three basic needs will be met, then we’re not doing our job as fellow Americans.”

The foundation and the city recently announced that the program, which was slated to complete this month, will run through June. Shah and Ambrosino said they hope to continue until summer.

Although the program appears to be helping, the need remains great in Chelsea. Liebman said that about 40 percent of people who applied for the program say they or their children don’t get enough to eat at times, compared with 55 percent before Chelsea Eats started.

At a Salvation Army pantry on Chestnut Street twice a week, Captain Brenda Gonzalez said people need groceries and other supplies despite the Chelsea Eats program. On Thursday, volunteers distributed turkey, fruits, vegetables, rice, toiletries, and other goods.

Many people who requested cards were not selected to receive them. The payments were distributed among 3,615 applicants by lottery. Others have needs that are beyond or outside of the program.

For the participants, the additional money allowed some financial flexibility. Several Chelsea Eats participants, who volunteered with the Salvation Army on Thursday, said they used the money for food, but also for cleaning supplies and clothing.

“The card was very helpful because it covered so many things that grocery brands didn’t,” said Gloria Caballero, who lost her job in the food service industry more than a year ago. The mother of two spoke to Globus in an interview interpreted by Gonzalez.

While the Harvard study didn’t pinpoint exactly what people bought with their cards, it segmented purchases based on the type of retailer at which transactions were made.

The results showed that 20 percent of spending was with retailers who primarily sell products other than groceries. Four percent went to what researchers called “professional services” – mostly utilities.

One percent went for travel and transportation. Less than half of 1 percent was spent in liquor stores or smoke shops.

While the near-term outlook for Chelsea Eats remains promising, Ambrosino hopes this can provide additional boost to basic income programs.

Ambrosino said widespread acceptance for such programs, including federal support, will be essential. “The local government cannot do this alone,” he said.

Andy Rosen can be reached at andrew.rosen@globe.com. Follow him on Twitter @andyrosen.